A TDM quasi-duplex system

Conventional two-way radio systems operating in the duplex mode require two different frequencies for communication in two directions to take place simultaneously. However, this wastes spectrum. The main attraction for any full-duplex two-way radio system is the ability for the listener to interrupt and speak at any time. An approach, illustrated in U.S. Pat. No. 4,037,158, and assigned to the same assignee as the instant application, is to provide simulated duplex operation on a single channel, utilizing hardware control of the transmitter and receiver by sampling the channel for the presence of a carrier wave. This approach generally requires increased radio complexity.

Time division multiplexing (TDM) of voice and data onto a single channel is known to provide full-duplex radio operation. Typically, however, this requires at least two time slots for a radio to transmit half the time and to receive at half the time. While effective, this practice wastes spectrum since, in ordinary conversations, the talker usually receives only brief feedback responses (or acknowledgements) during his or her transmissions.

Accordingly, it is an object of the present invention to provide an improved quasi-duplex time division multiplexed communication system.

Briefly, according to the invention, the primary feature of full-duplex communications is provided while conserving spectrum and without undue complexity. A talker transmits while a listener receivers until a command (interrupt) code is transmitted by the listener. After appropriate signalling, the talker and listener changes roles and the original talker now listens. Preferably, the time allocated for voice transmission is greater than the time allocateed for receiving data for system synchronization or commands to maximize speech capacity throughput.

FIG. 1 is a block diagram of a communication system utilizing a TDM Quasi-Duplex System in accordance with the present invention.

FIG. 2 is a block diagram of either radio of FIG. 1.

FIGS. 3a-m are timing diagrams illustrating the operation of the system of FIG. 1.

Referring now by characters of reference to the drawings, and first to FIG. 1, a communication system utilizing a TDM Quasi-Duplex System in accordance with the present invention is illustrated. Basically, a two-way radio (or station) 10 includes a transmitter 12, a receiver 14, a controller 20, and an antenna switch 18. The transmitter 12 and receiver 14 are selecetively connected to an antenna 16 via the switch 18. A manually operated key 31 coupled to one input of the controller 20, or a voice operated key 29 coupled to one input of the transmitter 12 selectively operates the controller 20. The radio 10 communicates with a second radio (or station) 22 having an antenna 24. The radio 22 may be identical to the radio 10 and may be connected to the antenna 24 as by an antenna switch 18 operated by a controller. As will be further explained, the controller 20, may include a microcomputer, which controls the operation of the transmitter 12, the receiver 14, and the switch 18.

FIG. 2 illustrates a block diagram of the radio 10 (or 22). The transmitter 12 comprises a microphone 28, an analog to digital (A/D) converter 32, a transmitter buffer 34, a digital to analog (D/A) converter 35, a modulator 36, and a power amplifier 38. The microphone 28 is coupled, via the A/D converter 32, to the controller 20, which controls the input to the transmitter buffer 34. The ouput of the transmitter buffer 34 is fed into the controller 20, which regulates the input to the D/A converter 35. The output of the D/A 35 is fed into the modulator 36, which is connected to the power amplifier 38. The controller 20 also connects to one input of the modulator 36, and to the lower amplifier 38 to gate the modulator bursts on and off. Lastly, as previously mentioned, the controller 20 is coupled to and controls the switch 18. The output of the power amplifier 38 is connected to the switch 18 for application to the antenna 16. These elements may be of conventional construction as commonly used in the radio arts.

The receiver 14 typically comprises a front end 39, an analog to digital (A/D) converter 41, a receiver buffer 42, a digital to analog (D/A) converter 44, and a speaker 46. The front end 39 is connected, via the analog to digital (A/D) converter 41, to the controller 20, which controls the input to the receiver buffer 42. The output of the receiver buffer 42 is applied to one input of the controller 20 for digitall compression before application to the input of the D/A converter 44, which in turn is connected to the speaker 46. The controller 20 also couples to the front end 39.

Operationally, audio energy impressed upon the microphone 28 is digitized at normal speed (32) and buffered in the transmitter buffer 34. To achieve time compression suitable to properly TDM the communication channel, information is retrieved from the transmit buffer 34 at a rate exceeding that of which it was stored. The duration and timing of the slots are defined in accordance with any suitable TDM protocol. The controller 20 forwards this information to the modulator 36, via the D/A converter 35, and the modulated information is broadcast via the power amplifier 38 and the antenna 16, after application of the switch 18. Symmetrically, received audio is sampled in the A/D converter 41 and stored at the higher rate in the receiver buffer 42. The contents of the buffer 42 are applied via the D/A at normal speed to the speaker to generate speaker audio.

While an embodiment of the invention utilizing an analog transmission has been described and shown, it should be understood that a digital transmission embodiment is also envisioned as is illustrated in U.S. Pat. No. 4,754,450 incorporated herein, and assigned to the same assignee as the instant application. In that invention as is here, voice signals for transmission are analyzed and vo-coded into a digital signal to be synthesized back to voice messages in the receiver.

In the digital embodiment, at the talker end, speech samples from an A/D converter would be processed continuously by a speech encoder (if required to reduce the speech information rate) and stored in a transmit buffer. During a voice transmit burst, speech data would be retrieved from the transmit buffer and transmitted at the channel rate (greater than the speech information rate). For example, if the speech information rate was 9600 BPS, a channel transmission rate of 12000 BPS would allow for four fifths of a frame to be split between transmission from talker to listener and one fifth for the sync data transmission from listener to talker and the transmit/receive gaps required for switching as is known in the art. At the listener end, speech data would be written into the receive buffer at the transmission rate during the receive burst and read from the buffer continuously to provide input to the speech decoder which in turn would provide normal speed voice to the speaker circuits.

Connected to the input of the controller 20 is the transmit key 31. This may comprise a push-to-talk (PTT) switch as is commonly used with a radio, and the key 31 and microphone may be provided as a single unit. Alternatively the transmit key may be a voice operated relay (VOX) 29 coupled to the output of the A/D converter 32. In either case the transmit key can be of known construction. Conventional voice activated transmit systems often loose the initial portion of the first utterance. However, this invention overcomes this problem since the mike audio is sampled and buffered by 32 and 34 respectively, as soon as the information is outputted at the mike 28. A pointer inside the controller 20 could be "backed up" at the beginning of the voice prior to initiating voice transmission, thereby avoiding loss of speech due to the detect delay in the receiver. Since the VOX delay time is known, it may be compensated by retrieving the speech samples starting from this known time ahead of the VOX activation.

The sequence of operations just described is illustrated in FIGS. 3a-m wherein a radio frequency communication channel 50 having a conventional bandwidth is apportioned into substantially larger voice slots 52 than data slots 54 to achieve Time Division Multiplexing (TDM) of voice and data on a single channel. Gaps (or guards bands) 56 between the slots are reserved for the proper operation of the switch 18.

As indicated when the key is manually (or voice) activated by a talker on the calling radio 10, the ID of the listener's radio to be called, 22, will be repeatedly transmitted in the voice slots 58 and 78. The protocol is illustrated for talker and listener together and separately for clarification. The called unit 22 detects the address and begins acquiring synchronization adequate for TDM. When synchronization has been achieved, the called unit 22 sends a data burst designated as SYNC in data slots 62 and 82. Detection of SYNC by the calling unit 10 signals the start of voice conversation between the talker 10 and the listener 22. The talker transmits and the listener receives voice during the voice slots 52 and 83, while the talker receives and the listener transmits data in the data slots 54 and 74.

To interrupt the talker 20 (i.e. to change roles), the listener 22 transmits a command, referred to as the turnaround request code (TREQ) in the data slots 64 and 84 by manually (or vocally) activating the key 31 (or 29) as indicated. When the talker 10 detects TREQ 64 and 84, it acknowledges and transmits an ACK word in the first portion of the former voice slots 66 and 86. Referring to FIGS. 3h-n the reception of TREQ 64 and 84 by the talker 10 overrides any further voice transmission and forces the talker into the role of the "new listener". To become the talker once more, key 31 must be released and re-activated. The previous listener 22 now is the "new talker" that will transmit voice from unit 22 to unit 10 in the subsequent voice slots 68 and 88, and the previous talker 10 now is the "new listener" who will transmit the SYNC word in the subsequent data slots 72 and 92 to maintain TDM synchronization.

For simplicity, only one talker per time slot is illustrated in the previous diagrams. However, this invention extends to a Time Division Multiple Access (TDMA) quasi-duplex system where multiple talkers transmit in subdivided time slots 92, 94 and receive in slots 96, 98. By providing SYNC words 102, 104 along with at least gaps 106, 108 necessary to ensure non-overlapping segments, time isolation may be maintained to allow for proper system synchronization.